论文部分内容阅读
从总体拟合效果、与疲劳物理的一致性和尾部预测的安全性角度，研究了可能的４种假设分布（三参数Ｗｅｉｂｕｌ、两参数Ｗｅｉｂｕｌ、对数正态和极大值分布）对管道结构焊接试样２４０℃低周疲劳裂纹萌生寿命数据的拟合效果。结果表明：三参数Ｗｅｉｂｕｌ分布总体拟合效果最好，但由于其形状参量可能出现小于１的情况，不符合疲劳物理要求；其尾部大多数情况可能给出非保守估计，不是一种良好的假设分布。其它三种分布的总体拟合效果稍差，但在工程应用范围内符合疲劳物理要求，尾部大多数情况可给出偏于保守的估计。因此，这３种假设分布可作为该接头疲劳可靠性分析的良好假设分布。而且，极大值分布与该焊接头的疲劳物理行为有良好的一致性。
From the perspective of the overall fitting effect, the consistency with the fatigue physics and the safety of the tail forecasting, the possible four kinds of hypothetical distributions (three-parameter Weibul, two-parameter Weibul, logarithmic normal and maximum distribution) Fitting Effect of Weld Specimen 240 ℃ Low Fatigue Crack Initiation Life Data. The results show that the three-parameter Weibul distribution has the best overall fitting result, but its shape parameters may not meet the fatigue physical requirements because their shape parameters may be less than 1. The non-conservative estimation may be given in most of the cases, which is not a good assumption distributed. The other three distributions have a slightly poorer overall fitting effect, but they meet the fatigue physics requirements within engineering applications, with the more conservative estimate given in most cases. Therefore, these three hypothetical distributions can be used as a good hypothetical distribution of the fatigue reliability analysis of the joint. Moreover, the maximum value distribution is in good agreement with the fatigue physical behavior of the welded joint.